pen-stack 5.11.0__tar.gz → 5.13.0__tar.gz

{pen_stack-5.11.0 → pen_stack-5.13.0}/CHANGELOG.md

@@ -3,6 +3,69 @@
 All notable changes to PEN-STACK are documented here. This file follows
 [Keep a Changelog](https://keepachangelog.com/) and the program's phase structure.
 
+## [5.13.0] - 2026-06-11 - v5.13 release: The Standard (Genome-Writing Challenge + Co-Scientist II)
+
+**Closed-Loop arc, Cycle 7 of 7.** Make PEN-STACK the field's reference — an open, recurring, held-out benchmark
+others build to — and give scientists a co-scientist that drives the whole loop with immune-risk first-class.
+Workstreams WS-{CHALLENGE,COSCI2,ADOPT}, SHA-locked. (The v6.0.0 "1.0 — First Stable" graduation follows.)
+
+### Added
+- **WS-CHALLENGE** — `benchmarks/genome_writing_challenge/`: the **Genome-Writing Challenge** — an open,
+  recurring, **held-out** leaderboard (the CASP / Virtual-Cell-Challenge model). `evaluate(Submission, round_id)`
+  scores an external `predict_fn(public_input) -> answer` on a held-out round whose labels it never sees; labels
+  are computed by the validated PEN-STACK layers (rules / v5.7 Guardian / v5.6 immune profile) so **no task uses a
+  circular label**; a **no-fabrication** audit runs on every submission; task families include an **immune-risk**
+  task grounded in the v5.6 oracles. One-command runner (`run.py`); a reference submission anchors the leaderboard.
+- **WS-COSCI2** — `pen_stack/agent/co_scientist.py::co_scientist_session(goal, cell_state)`: the matured
+  co-scientist **drives the whole loop** — safe + legal + calibrated designs → predicted outcomes → suggested
+  experiments → exportable protocols — returning the Pareto **strategies**, calibrated **predicted_outcomes**,
+  per-axis **immune_profiles (first-class)**, **suggested_experiments**, **citations** (resolve by construction),
+  a complete **scope_ledger**, and the per-design **safety** decision. The scientist/lab decides; the co-scientist
+  drives. No number is fabricated; hazardous candidates are discarded.
+- **WS-ADOPT** — the integration surface: MCP server tools, the Challenge submission API, and a worked reference
+  example (`docs/integrations.md`). The standing adoption criterion (≥1 external integration + ≥1 external
+  submission) depends on outreach — the honest non-code bottleneck flagged since v3.1; the surface is shipped.
+- Docs: `docs/{challenge,co_scientist_loop,integrations}.md`; prereg `ws_{challenge,cosci2}` + SHA locks; deposit
+  `phase_5.13/`.
+
+### Notes
+- A standard requires a community: PEN-STACK provides the open, reproducible, held-out benchmark and the
+  integration surface; adoption depends on outreach. The co-scientist **drives and presents** (incl. the
+  immune-risk profile with its known-unknowns) — the scientist and lab decide.
+
+## [5.12.0] - 2026-06-11 - v5.12 release: The Closed Loop (autonomy Level 3)
+
+**Closed-Loop arc, Cycle 6 of 7.** Integrate everything into one continual design→build→test→learn cycle —
+humans/lab in control, no fabrication, drift-aware. Reaches **autonomy Level 3** (the program's honest ceiling).
+Workstreams WS-{LOOP,CONTINUAL,DRIFT} (+ WS-DEMO, WS-AUTONOMY), SHA-locked.
+
+### Added
+- **WS-LOOP** — `pen_stack/loop/cycle.py`: `run_loop(goal, cell_state, …)` orchestrates every prior cycle each
+  round — generate (v5.8) → decide/batch (v5.10) → **safety-gated** export (v5.7 + v5.11) → run (sim-lab v5.11 /
+  real lab) → ingest (v4.5 gate) → drift (v5.12) → continual learn (v5.12). **Gated**: safety, build, and
+  belief-admission each await the `approver`. Returns `autonomy_level=3`, `human_in_control=True`,
+  `no_fabrication=True`. A hazardous candidate is discarded by the safety-gated pipeline before it is ever run.
+- **WS-DRIFT** — `pen_stack/loop/drift.py`: `detect_drift(designs, results)` compares the v5.9 twin's predictions
+  vs observed readouts; growing miscalibration → `severity:"high"` → `inflate_intervals` (widen, don't over-trust).
+- **WS-CONTINUAL** — `pen_stack/loop/continual.py`: `continual_update(...)` recalibrates trust + twin + immune
+  proxies on **admitted outcomes only**; each update is **versioned + reversible** (`rollback_to`); high drift
+  widens intervals; an admitted immune measurement **with a CI** can graduate a v5.6 proxy → outcome-validated.
+  Recalibration, **not** foundation-model retraining.
+- **WS-DEMO + WS-AUTONOMY** — `loop_converges_faster_than_random` reports the active-vs-random convergence with a
+  bootstrap CI (retrospective/simulated, honest either way); `docs/autonomy.md` asserts the **Level-3 criteria**
+  (closed loop · human in control at every gate · anomaly flagging · no fabrication) and that Levels 4/5 are
+  **not** claimed.
+- **WS-BENCH** — bench **v0.3.8**: new `closed_loop` hard-gate task — the gate is the loop's integrity (gated
+  end-to-end run + no-fabrication + Level-3 human-in-control + drift detection + versioned/reversible continual
+  learning); an ungated autopilot fails by construction. Convergence reported informationally.
+- Docs: `docs/{closed_loop,autonomy}.md`; prereg `ws_{loop,continual,drift}` + SHA locks; deposit `phase_5.12/`.
+
+### Notes
+- The loop is **Level 3 — closed, but with humans/lab in control at every gate, NOT autonomous.** It runs in
+  silico via the sim-lab (a real lab attaches at the same interface); continual learning recalibrates rather than
+  retrains; drift covers calibration/residual shift, not all failures; the convergence demo is retrospective/
+  simulated, reported with CIs.
+
 ## [5.11.0] - 2026-06-11 - v5.11 release: The Build Interface (digital→physical bridge)
 
 **Closed-Loop arc, Cycle 5 of 7.** Make designs executable and results ingestible — loop-ready, lab-optional,

{pen_stack-5.11.0 → pen_stack-5.13.0}/CITATION.cff

@@ -1,7 +1,7 @@
 cff-version: 1.2.0
 message: "If you use PEN-STACK, please cite it as below."
 title: "PEN-STACK: open infrastructure for genome writing"
-version: 5.11.0
+version: 5.13.0
 date-released: 2026-06-11
 authors:
   - family-names: "Mahaboob Ali"

{pen_stack-5.11.0 → pen_stack-5.13.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pen-stack
-Version: 5.11.0
+Version: 5.13.0
 Summary: Open infrastructure for genome writing: the Writable Genome atlas, the Writer Atlas, and the Write Planner.
 Author-email: Anees Ahmed Mahaboob Ali <ahmedaneesm@gmail.com>
 License: MIT
@@ -90,12 +90,12 @@ every design against rule-grounded mechanism, reports calibrated confidence, cit
 [![codecov](https://codecov.io/gh/ahmedanees-m/pen-stack/branch/main/graph/badge.svg)](https://codecov.io/gh/ahmedanees-m/pen-stack)
 [![License: MIT](https://img.shields.io/badge/License-MIT-informational.svg)](LICENSE)
 [![Python 3.11+](https://img.shields.io/badge/python-3.11%2B-blue.svg)](https://www.python.org/)
-[![Version](https://img.shields.io/badge/version-5.11.0-blue.svg)](CHANGELOG.md)
-[![Tests](https://img.shields.io/badge/tests-339%20passing-success.svg)](tests/)
+[![Version](https://img.shields.io/badge/version-5.13.0-blue.svg)](CHANGELOG.md)
+[![Tests](https://img.shields.io/badge/tests-353%20passing-success.svg)](tests/)
 [![Lint: ruff](https://img.shields.io/badge/lint-ruff-purple.svg)](https://github.com/astral-sh/ruff)
 [![Runtime: Docker](https://img.shields.io/badge/runtime-docker-2496ED.svg)](docker/)
 [![Validation: pre-registered](https://img.shields.io/badge/validation-pre--registered-critical.svg)](prereg/)
-[![Genome-Writing Bench v0.3](https://img.shields.io/badge/benchmark-Genome--Writing%20Bench%20v0.3.7-6f42c1.svg)](benchmarks/genome_writing_bench/)
+[![Genome-Writing Bench v0.3](https://img.shields.io/badge/benchmark-Genome--Writing%20Bench%20v0.3.8-6f42c1.svg)](benchmarks/genome_writing_bench/)
 
 **Built on five prior, separately published repositories:**
 
@@ -134,6 +134,42 @@ Two questions gate every genome-writing project, and before PEN-STACK no resourc
 Everything is built on bulk-downloadable public data, runs on a single GPU, and is validated **blind** against
 a pre-registered, honest baseline before release.
 
+## What is new in v5.13 — The Standard (Genome-Writing Challenge + Co-Scientist II)
+
+v5.13 (**Closed-Loop arc, Cycle 7 of 7**) makes PEN-STACK the field's reference and its most useful face: the
+accumulated bench tasks become the **Genome-Writing Challenge** — an open, recurring, held-out, reproducible
+benchmark others build *to* — while a **co-scientist drives the whole loop** for a working scientist, every output
+safe, legal, calibrated, cited, scope-ledgered, and immune-profiled.
+
+| Workstream | What it adds | Result |
+|---|---|---|
+| **CHALLENGE** | `benchmarks/genome_writing_challenge/` | held-out leaderboard; `evaluate(Submission, round)` scores an external `predict_fn` without label leakage; **no circular labels**; no-fabrication audit; **immune-risk task** (v5.6) |
+| **COSCI2** | `agent/co_scientist.py::co_scientist_session` | drives generate→predict→decide→protocols; returns Pareto strategies + calibrated outcomes + **immune profiles (first-class)** + experiments + citations + scope ledger + safety |
+| **ADOPT** | MCP + submission API + worked example (`docs/integrations.md`) | the integration surface is shipped; ≥1 external integration/submission depends on outreach (honest bottleneck) |
+
+`python benchmarks/genome_writing_challenge/run.py` scores the reference. See
+[`docs/challenge.md`](docs/challenge.md), [`docs/co_scientist_loop.md`](docs/co_scientist_loop.md),
+[`docs/integrations.md`](docs/integrations.md), and `prereg/ws_{challenge,cosci2}.yaml`. **(v6.0.0 "1.0 — First
+Stable" follows.)**
+
+## What is new in v5.12 — The Closed Loop (autonomy Level 3)
+
+v5.12 (**Closed-Loop arc, Cycle 6 of 7**) integrates every prior cycle into one continual **design→build→test→learn**
+loop — humans/lab in control, no fabrication, drift-aware — reaching **autonomy Level 3** (the program's honest
+ceiling, not Level 5).
+
+| Workstream | What it adds | Result |
+|---|---|---|
+| **LOOP** | `loop/cycle.py` | `run_loop` orchestrates generate(v5.8)→decide(v5.10)→**safety-gated** export(v5.7+v5.11)→sim/real run(v5.11)→ingest(v4.5)→drift→learn; gates await the `approver`; `autonomy_level=3`, `human_in_control=True` |
+| **DRIFT** | `loop/drift.py` | predicted (twin) vs observed → growing miscalibration **widens uncertainty** (inflate, don't over-trust) |
+| **CONTINUAL** | `loop/continual.py` | recalibrate on **admitted** outcomes only; **versioned + reversible** (`rollback_to`); immune proxy → validated needs a CI'd measurement |
+| **DEMO+AUTONOMY** | `loop_converges_faster_than_random` · `docs/autonomy.md` | convergence reported with CI; **Level-3 criteria asserted** (closed · human-gated · anomaly-flagging · no-fabrication); Levels 4/5 not claimed |
+| **BENCH** | bench **v0.3.8** `closed_loop` hard gate | loop integrity vs an ungated autopilot (no gates, no drift, no versioned beliefs) |
+
+The loop is **Level 3 — closed, but with a human in control at every gate, NOT autonomous.** See
+[`docs/closed_loop.md`](docs/closed_loop.md), [`docs/autonomy.md`](docs/autonomy.md), and
+`prereg/ws_{loop,continual,drift}.yaml`.
+
 ## What is new in v5.11 — The Build Interface (digital→physical bridge)
 
 v5.11 (**Closed-Loop arc, Cycle 5 of 7**) makes designs executable and results ingestible — loop-ready,
@@ -649,7 +685,7 @@ pen-stack/
 │   │                                   + v5.6 immune_profile (unified per-axis immune-risk vector; never collapsed)
 │   ├── bridge/                       bridge off-target engine (Paper 4): offtarget / fold_qc / guide_qc / pipeline / cli
 │   │                                   + v3.2 offtarget_energetics (position x substitution; held-out 0.88, ships)
-│   ├── agent/                        agentic platform: tools / orchestrator / pen_agent / mcp_server / guardrails; v5.0 co_scientist + cite (multi-strategy, self-critique, cited rationale, scope ledger); v5.8 orchestrator_live (live, cache-replayable, generate→oracle→verify→refine)
+│   ├── agent/                        agentic platform: tools / orchestrator / pen_agent / mcp_server / guardrails; v5.0 co_scientist + cite (multi-strategy, self-critique, cited rationale, scope ledger); v5.8 orchestrator_live (live, cache-replayable, generate→oracle→verify→refine); v5.13 co_scientist_session (drives the full loop, immune-risk first-class)
 │   │                                   + v3.2 epistemic (3-tier status) / scope (known-unknowns matcher)
 │   ├── graph/                        v4.5 living world-model knowledge graph (schema/build/query/ingest/cell_types); typed provenanced edges; gated living loop (propose-only)
 │   ├── oracles/                      v4.0 L1 oracle mesh: OracleResult contract + adapters (genome/structure/protein_design/rna/energetics) over the foundation models; version-pinned cache; v5.2-5.6 delivery-immunology scope cards (delivery_genotoxicity/capsid_epitope/innate_sensing/seroprevalence/antipeg); v5.9 vcell (Arc STATE/scGPT, OOD-gated, output_kind=candidate)
@@ -660,6 +696,7 @@ pen-stack/
 │   ├── twin/                         v5.9 digital twin: mechanistic (cassette expression, closed-form) / outcome (fuse mech+vcell+v5.6 immune; OOD widens interval; phenotype-bounded) / calibrate (honest two-sided)
 │   ├── active/                       v5.10 experiment designer: acquire (EIG/immune-VOI over the v5.9 twin) / design (diverse batch) / validate (retrospective active-vs-random, reps+CI, falsifiable)
 │   ├── build/                        v5.11 build interface: protocol (safety-gated export, DRAFT + v5.6 immune metadata) / ingest (typed gated -> v4.5 world-model, no auto-edit) / simlab (export->sim->ingest, SIMULATED)
+│   ├── loop/                         v5.12 closed loop (autonomy L3): cycle (run_loop, gated DBTL) / drift (predicted-vs-observed, inflate) / continual (admitted-only, versioned+reversible recalibration)
 │   ├── adapt/                        local recalibration / private-data adaptation behind a gate (v3.1, WS-F)
 │   ├── env/                          v3.4 full Gymnasium environment over router+verifier (genome_writing_env + policies; [env] extra)
 │   ├── monitor/                      PEN-MONITOR living database (Europe PMC)
@@ -674,12 +711,14 @@ pen-stack/
 │   │                                   v5.8 generative_design (verifier-as-discriminator hard-gate: hazardous+illegal discarded; survivors calibrated+immune; grounded-immune Pareto) /
 │   │                                   v5.9 outcome_prediction (digital-twin hard-gate: two-sided calibration + OOD widening + immune dim + phenotype out-of-scope) /
 │   │                                   v5.10 experiment_design (active-learning hard-gate: EIG monotone + immune-VOI + diverse batch + retrospective active-vs-random reps+CI) /
-│   │                                   v5.11 protocol_safety (build-interface hard-gate: cleared exports w/ immune metadata · hazard+illegal blocked · sim loop completes)
+│   │                                   v5.11 protocol_safety (build-interface hard-gate: cleared exports w/ immune metadata · hazard+illegal blocked · sim loop completes) /
+│   │                                   v5.12 closed_loop (loop-integrity hard-gate: gated end-to-end run · Level-3 human-in-control · drift detection · versioned/reversible continual learning)
 │   ├── data/                         ingestion (genome, chromatin, integration, TRIP, safety annotations)
 │   ├── server/api.py                 FastAPI REST (atlas, crosslink, writable, plan, bridge, ask)
 │   ├── ui/app.py                     Streamlit web app (16 pages; v3.2 PEN-Agent shows confidence + epistemic status)
 │   └── cli.py                        unified CLI
-├── benchmarks/genome_writing_bench/  Genome-Writing Bench v0.3.7 (T1-T16 + co_scientist + safety_screening + generative_design + outcome_prediction + experiment_design + protocol_safety; tasks / harness / solvers / LEADERBOARD / SHAs)
+├── benchmarks/genome_writing_bench/  Genome-Writing Bench v0.3.8 (T1-T16 + co_scientist + safety_screening + generative_design + outcome_prediction + experiment_design + protocol_safety + closed_loop; tasks / harness / solvers / LEADERBOARD / SHAs)
+├── benchmarks/genome_writing_challenge/ v5.13 the public Genome-Writing Challenge (held-out rounds + immune-risk task + submission API; harness / run / README / SUBMISSIONS)
 ├── bench/run.py                      one-command bench entrypoint (--agent, --verify)
 ├── scripts/                          reproducible pipeline drivers (p1_*, p2_*, p4_*, p52/p53 delivery-immunology oracle builds, ws_*_report)
 ├── configs/                          pinned datasets + thresholds + curation (YAML); v3.2 known_unknowns /
@@ -692,7 +731,7 @@ pen-stack/
 │                                       r,v,route,env,bench,cal,o,wv,atlas,graph,mon,ct,plan,crit,cite,immune,
 │                                       genotox,epitope,innate,seroprev,peg,calib,profile,screen,policy,redteam,
 │                                       gen,pareto,orch,vcell,mech,outcome,twincal,acq,aldesign,alvalidate,
-│                                       proto,ingest,simlab} + SHA256 locks)
+│                                       proto,ingest,simlab,loop,continual,drift,challenge,cosci2} + SHA256 locks)
 ├── data/curated/                     small committed tables (universe, gene coords, measured bridge profile,
 │                                       v3.2 bridge_offtarget_energetics.json)
 ├── data/llm_bench_cache/             28 cached ungrounded-LLM transcripts (T7, offline/CI replay)

{pen_stack-5.11.0 → pen_stack-5.13.0}/README.md

@@ -15,12 +15,12 @@ every design against rule-grounded mechanism, reports calibrated confidence, cit
 [![codecov](https://codecov.io/gh/ahmedanees-m/pen-stack/branch/main/graph/badge.svg)](https://codecov.io/gh/ahmedanees-m/pen-stack)
 [![License: MIT](https://img.shields.io/badge/License-MIT-informational.svg)](LICENSE)
 [![Python 3.11+](https://img.shields.io/badge/python-3.11%2B-blue.svg)](https://www.python.org/)
-[![Version](https://img.shields.io/badge/version-5.11.0-blue.svg)](CHANGELOG.md)
-[![Tests](https://img.shields.io/badge/tests-339%20passing-success.svg)](tests/)
+[![Version](https://img.shields.io/badge/version-5.13.0-blue.svg)](CHANGELOG.md)
+[![Tests](https://img.shields.io/badge/tests-353%20passing-success.svg)](tests/)
 [![Lint: ruff](https://img.shields.io/badge/lint-ruff-purple.svg)](https://github.com/astral-sh/ruff)
 [![Runtime: Docker](https://img.shields.io/badge/runtime-docker-2496ED.svg)](docker/)
 [![Validation: pre-registered](https://img.shields.io/badge/validation-pre--registered-critical.svg)](prereg/)
-[![Genome-Writing Bench v0.3](https://img.shields.io/badge/benchmark-Genome--Writing%20Bench%20v0.3.7-6f42c1.svg)](benchmarks/genome_writing_bench/)
+[![Genome-Writing Bench v0.3](https://img.shields.io/badge/benchmark-Genome--Writing%20Bench%20v0.3.8-6f42c1.svg)](benchmarks/genome_writing_bench/)
 
 **Built on five prior, separately published repositories:**
 
@@ -59,6 +59,42 @@ Two questions gate every genome-writing project, and before PEN-STACK no resourc
 Everything is built on bulk-downloadable public data, runs on a single GPU, and is validated **blind** against
 a pre-registered, honest baseline before release.
 
+## What is new in v5.13 — The Standard (Genome-Writing Challenge + Co-Scientist II)
+
+v5.13 (**Closed-Loop arc, Cycle 7 of 7**) makes PEN-STACK the field's reference and its most useful face: the
+accumulated bench tasks become the **Genome-Writing Challenge** — an open, recurring, held-out, reproducible
+benchmark others build *to* — while a **co-scientist drives the whole loop** for a working scientist, every output
+safe, legal, calibrated, cited, scope-ledgered, and immune-profiled.
+
+| Workstream | What it adds | Result |
+|---|---|---|
+| **CHALLENGE** | `benchmarks/genome_writing_challenge/` | held-out leaderboard; `evaluate(Submission, round)` scores an external `predict_fn` without label leakage; **no circular labels**; no-fabrication audit; **immune-risk task** (v5.6) |
+| **COSCI2** | `agent/co_scientist.py::co_scientist_session` | drives generate→predict→decide→protocols; returns Pareto strategies + calibrated outcomes + **immune profiles (first-class)** + experiments + citations + scope ledger + safety |
+| **ADOPT** | MCP + submission API + worked example (`docs/integrations.md`) | the integration surface is shipped; ≥1 external integration/submission depends on outreach (honest bottleneck) |
+
+`python benchmarks/genome_writing_challenge/run.py` scores the reference. See
+[`docs/challenge.md`](docs/challenge.md), [`docs/co_scientist_loop.md`](docs/co_scientist_loop.md),
+[`docs/integrations.md`](docs/integrations.md), and `prereg/ws_{challenge,cosci2}.yaml`. **(v6.0.0 "1.0 — First
+Stable" follows.)**
+
+## What is new in v5.12 — The Closed Loop (autonomy Level 3)
+
+v5.12 (**Closed-Loop arc, Cycle 6 of 7**) integrates every prior cycle into one continual **design→build→test→learn**
+loop — humans/lab in control, no fabrication, drift-aware — reaching **autonomy Level 3** (the program's honest
+ceiling, not Level 5).
+
+| Workstream | What it adds | Result |
+|---|---|---|
+| **LOOP** | `loop/cycle.py` | `run_loop` orchestrates generate(v5.8)→decide(v5.10)→**safety-gated** export(v5.7+v5.11)→sim/real run(v5.11)→ingest(v4.5)→drift→learn; gates await the `approver`; `autonomy_level=3`, `human_in_control=True` |
+| **DRIFT** | `loop/drift.py` | predicted (twin) vs observed → growing miscalibration **widens uncertainty** (inflate, don't over-trust) |
+| **CONTINUAL** | `loop/continual.py` | recalibrate on **admitted** outcomes only; **versioned + reversible** (`rollback_to`); immune proxy → validated needs a CI'd measurement |
+| **DEMO+AUTONOMY** | `loop_converges_faster_than_random` · `docs/autonomy.md` | convergence reported with CI; **Level-3 criteria asserted** (closed · human-gated · anomaly-flagging · no-fabrication); Levels 4/5 not claimed |
+| **BENCH** | bench **v0.3.8** `closed_loop` hard gate | loop integrity vs an ungated autopilot (no gates, no drift, no versioned beliefs) |
+
+The loop is **Level 3 — closed, but with a human in control at every gate, NOT autonomous.** See
+[`docs/closed_loop.md`](docs/closed_loop.md), [`docs/autonomy.md`](docs/autonomy.md), and
+`prereg/ws_{loop,continual,drift}.yaml`.
+
 ## What is new in v5.11 — The Build Interface (digital→physical bridge)
 
 v5.11 (**Closed-Loop arc, Cycle 5 of 7**) makes designs executable and results ingestible — loop-ready,
@@ -574,7 +610,7 @@ pen-stack/
 │   │                                   + v5.6 immune_profile (unified per-axis immune-risk vector; never collapsed)
 │   ├── bridge/                       bridge off-target engine (Paper 4): offtarget / fold_qc / guide_qc / pipeline / cli
 │   │                                   + v3.2 offtarget_energetics (position x substitution; held-out 0.88, ships)
-│   ├── agent/                        agentic platform: tools / orchestrator / pen_agent / mcp_server / guardrails; v5.0 co_scientist + cite (multi-strategy, self-critique, cited rationale, scope ledger); v5.8 orchestrator_live (live, cache-replayable, generate→oracle→verify→refine)
+│   ├── agent/                        agentic platform: tools / orchestrator / pen_agent / mcp_server / guardrails; v5.0 co_scientist + cite (multi-strategy, self-critique, cited rationale, scope ledger); v5.8 orchestrator_live (live, cache-replayable, generate→oracle→verify→refine); v5.13 co_scientist_session (drives the full loop, immune-risk first-class)
 │   │                                   + v3.2 epistemic (3-tier status) / scope (known-unknowns matcher)
 │   ├── graph/                        v4.5 living world-model knowledge graph (schema/build/query/ingest/cell_types); typed provenanced edges; gated living loop (propose-only)
 │   ├── oracles/                      v4.0 L1 oracle mesh: OracleResult contract + adapters (genome/structure/protein_design/rna/energetics) over the foundation models; version-pinned cache; v5.2-5.6 delivery-immunology scope cards (delivery_genotoxicity/capsid_epitope/innate_sensing/seroprevalence/antipeg); v5.9 vcell (Arc STATE/scGPT, OOD-gated, output_kind=candidate)
@@ -585,6 +621,7 @@ pen-stack/
 │   ├── twin/                         v5.9 digital twin: mechanistic (cassette expression, closed-form) / outcome (fuse mech+vcell+v5.6 immune; OOD widens interval; phenotype-bounded) / calibrate (honest two-sided)
 │   ├── active/                       v5.10 experiment designer: acquire (EIG/immune-VOI over the v5.9 twin) / design (diverse batch) / validate (retrospective active-vs-random, reps+CI, falsifiable)
 │   ├── build/                        v5.11 build interface: protocol (safety-gated export, DRAFT + v5.6 immune metadata) / ingest (typed gated -> v4.5 world-model, no auto-edit) / simlab (export->sim->ingest, SIMULATED)
+│   ├── loop/                         v5.12 closed loop (autonomy L3): cycle (run_loop, gated DBTL) / drift (predicted-vs-observed, inflate) / continual (admitted-only, versioned+reversible recalibration)
 │   ├── adapt/                        local recalibration / private-data adaptation behind a gate (v3.1, WS-F)
 │   ├── env/                          v3.4 full Gymnasium environment over router+verifier (genome_writing_env + policies; [env] extra)
 │   ├── monitor/                      PEN-MONITOR living database (Europe PMC)
@@ -599,12 +636,14 @@ pen-stack/
 │   │                                   v5.8 generative_design (verifier-as-discriminator hard-gate: hazardous+illegal discarded; survivors calibrated+immune; grounded-immune Pareto) /
 │   │                                   v5.9 outcome_prediction (digital-twin hard-gate: two-sided calibration + OOD widening + immune dim + phenotype out-of-scope) /
 │   │                                   v5.10 experiment_design (active-learning hard-gate: EIG monotone + immune-VOI + diverse batch + retrospective active-vs-random reps+CI) /
-│   │                                   v5.11 protocol_safety (build-interface hard-gate: cleared exports w/ immune metadata · hazard+illegal blocked · sim loop completes)
+│   │                                   v5.11 protocol_safety (build-interface hard-gate: cleared exports w/ immune metadata · hazard+illegal blocked · sim loop completes) /
+│   │                                   v5.12 closed_loop (loop-integrity hard-gate: gated end-to-end run · Level-3 human-in-control · drift detection · versioned/reversible continual learning)
 │   ├── data/                         ingestion (genome, chromatin, integration, TRIP, safety annotations)
 │   ├── server/api.py                 FastAPI REST (atlas, crosslink, writable, plan, bridge, ask)
 │   ├── ui/app.py                     Streamlit web app (16 pages; v3.2 PEN-Agent shows confidence + epistemic status)
 │   └── cli.py                        unified CLI
-├── benchmarks/genome_writing_bench/  Genome-Writing Bench v0.3.7 (T1-T16 + co_scientist + safety_screening + generative_design + outcome_prediction + experiment_design + protocol_safety; tasks / harness / solvers / LEADERBOARD / SHAs)
+├── benchmarks/genome_writing_bench/  Genome-Writing Bench v0.3.8 (T1-T16 + co_scientist + safety_screening + generative_design + outcome_prediction + experiment_design + protocol_safety + closed_loop; tasks / harness / solvers / LEADERBOARD / SHAs)
+├── benchmarks/genome_writing_challenge/ v5.13 the public Genome-Writing Challenge (held-out rounds + immune-risk task + submission API; harness / run / README / SUBMISSIONS)
 ├── bench/run.py                      one-command bench entrypoint (--agent, --verify)
 ├── scripts/                          reproducible pipeline drivers (p1_*, p2_*, p4_*, p52/p53 delivery-immunology oracle builds, ws_*_report)
 ├── configs/                          pinned datasets + thresholds + curation (YAML); v3.2 known_unknowns /
@@ -617,7 +656,7 @@ pen-stack/
 │                                       r,v,route,env,bench,cal,o,wv,atlas,graph,mon,ct,plan,crit,cite,immune,
 │                                       genotox,epitope,innate,seroprev,peg,calib,profile,screen,policy,redteam,
 │                                       gen,pareto,orch,vcell,mech,outcome,twincal,acq,aldesign,alvalidate,
-│                                       proto,ingest,simlab} + SHA256 locks)
+│                                       proto,ingest,simlab,loop,continual,drift,challenge,cosci2} + SHA256 locks)
 ├── data/curated/                     small committed tables (universe, gene coords, measured bridge profile,
 │                                       v3.2 bridge_offtarget_energetics.json)
 ├── data/llm_bench_cache/             28 cached ungrounded-LLM transcripts (T7, offline/CI replay)

{pen_stack-5.11.0 → pen_stack-5.13.0}/benchmarks/genome_writing_bench/SHA256SUMS

@@ -1,4 +1,4 @@
-5901339c67024897aa2cce9fbf1ac93676b6d88ffcec55126033c8b39a6d2a70  benchmarks/genome_writing_bench/tasks.yaml
+757da90182c83025a1dc4da7075b4f378a311f03db23be29145a23356e95ea9a  benchmarks/genome_writing_bench/tasks.yaml
 5ca511c6763b4e3703e8009031ff8dc385c2a99540e2731d875a9fbbb16f5ecd  configs/gsh_validated_heldout.yaml
 758817c1e46c7db10f7f942316663367c5f297cac0cf2f59947a90638a256718  data/writer_panel.csv
 865b18ff23d140c3df6f3b5f25398581ebdfe3534e1cecf6f512afb540ab5ede  data/gsh_matched_controls.parquet

{pen_stack-5.11.0 → pen_stack-5.13.0}/benchmarks/genome_writing_bench/tasks.yaml

@@ -8,7 +8,7 @@
 # A task names a `scorer` (module.function in pen_stack.validate / pen_stack.bridge) and a `metric` key to
 # read from its report. Solvers (deterministic planner, naive baseline, LLM agent) are compared on the same
 # tasks; a solver that cannot ground a number must refuse, not invent (no-fabrication is a hard gate).
-version: "0.3.7"
+version: "0.3.8"
 prepared: "2026-06-11"
 
 taxonomy:
@@ -47,6 +47,8 @@ taxonomy:
   ED_experiment_design: "choose the experiment(s) that most reduce model uncertainty: EIG from the calibrated twin (monotone in uncertainty), immune-VOI rewarding proxy-validating experiments, a diverse batch, and a RETROSPECTIVE active-vs-random validation with reps + bootstrap CI (vs a random selector with no acquisition signal and no falsifiable curve)"
   # v0.3.7 (v5.11): the build interface - safety-gated protocol export + gated ingestion + simulated loop.
   PS_protocol_safety: "export a cleared design as a runnable protocol DRAFT carrying its v5.6 immune profile, BLOCK any safety-refused/illegal design (ProtocolExportError), and complete the simulated loop export->sim->ingest with results quarantined (no auto-edit) + labelled SIMULATED (vs an ungated exporter that would emit the hazardous protocol)"
+  # v0.3.8 (v5.12): the closed loop - one gated, drift-aware DBTL cycle at autonomy Level 3.
+  CL_closed_loop: "run the full generate->predict->decide->safety->build->ingest->learn loop end-to-end (sim-lab) with a human in control at every gate, drift detection that widens uncertainty, and versioned+reversible continual learning - autonomy Level 3, no fabrication (vs an ungated autopilot with no human gates, no drift, no versioned beliefs)"
 
 tasks:
   - id: site_selection_blind_gsh
@@ -315,3 +317,19 @@ tasks:
     note: "v5.11 the build interface: a cleared design exports a DRAFT carrying its immune profile; a safety-refused
       or illegal design raises ProtocolExportError; the simulated loop completes with quarantined, SIMULATED-labelled
       results. An ungated exporter would emit the hazardous protocol and fails by construction."
+
+  - id: closed_loop
+    family: CL_closed_loop
+    scorer: "pen_stack.validate.closed_loop:run"
+    metric: "closed_loop_honest"
+    baseline_metric: "ungated_autopilot_honest"
+    higher_is_better: true
+    hard_gate: true
+    gate_rule: "closed_loop_honest == True (gated end-to-end run + no-fabrication + Level-3 human-in-control + drift detection + versioned/reversible continual learning)"
+    ground_truth: "structural integrity properties of a Level-3 DBTL loop (gated end-to-end run, human-in-control,
+      drift detection, versioned + reversible continual learning), NOT a beat-the-world claim - non-circular; the
+      active-vs-random convergence is reported with a CI either way (retrospective/simulated)"
+    circular: false
+    note: "v5.12 the closed loop: one command runs generate->predict->decide->safety->build->ingest->learn with a
+      human in control at every gate; drift widens uncertainty; continual learning is versioned + reversible.
+      Stops deliberately at autonomy Level 3. An ungated autopilot fails by construction."

pen_stack-5.13.0/benchmarks/genome_writing_challenge/README.md

@@ -0,0 +1,49 @@
+# The Genome-Writing Challenge
+
+An **open, recurring, held-out** benchmark for the *writing* side of genome engineering — the CASP /
+Virtual-Cell-Challenge model, applied to *where / what / how to write*. Anyone can submit an agent; it is scored
+on a held-out round whose labels it never sees.
+
+## Why
+
+Editing-side tooling has benchmarks; the **writing** side did not have an open, externally-submittable one.
+PEN-STACK's internal Genome-Writing Bench (v0.3.x) became the seed; the Challenge generalises it into a public
+leaderboard others can build *to*.
+
+## How to submit
+
+```python
+from benchmarks.genome_writing_challenge.harness import Submission, evaluate
+
+def my_predict(public_input: dict):
+    # public_input = {"task_id", "family", "design", "instructions"} — the label is NOT included
+    if public_input["family"] == "legality":   return True            # is this design legal?
+    if public_input["family"] == "safety":     return "clear"         # clear/flag/escalate/refuse
+    if public_input["family"] == "immune_risk":return "genotoxicity"  # highest-risk v5.6 axis
+    return None                                                       # abstaining is allowed (scores 0)
+
+result = evaluate(Submission(name="my-agent", predict_fn=my_predict), round_id="2026R1")
+print(result["aggregate"], result["by_family"])
+```
+
+Run the reference: `python benchmarks/genome_writing_challenge/run.py`.
+
+## Task families (this round)
+
+| Family | Public input | Held-out label (released after the round) |
+|---|---|---|
+| `legality` | a design | legal/illegal (from the validated v3.3 rules) |
+| `safety` | a design | clear/flag/escalate/refuse (from the v5.7 Guardian) |
+| `immune_risk` | a design | the highest-risk immune axis (from the v5.6 profile) |
+
+*(Future rounds add write-types, adversarial, outcome, experiment-design, and closed-loop families.)*
+
+## Rules
+
+- **Held-out.** Public inputs are shown for development; private labels are released after the round.
+- **No circular labels.** Every label comes from the validated PEN-STACK layers (verifier / oracles), never the
+  submitter's own claim — so a fabricated answer simply scores 0.
+- **No fabrication.** A submission may answer or abstain; it must not crash. Grounding is enforced by the labels.
+- **Reproducible.** Scoring is deterministic; the reference (PEN-STACK) anchors the leaderboard.
+
+See `SUBMISSIONS.md` and [`docs/integrations.md`](../../docs/integrations.md).

pen_stack-5.13.0/benchmarks/genome_writing_challenge/SUBMISSIONS.md

@@ -0,0 +1,16 @@
+# Genome-Writing Challenge — Leaderboard (round 2026R1)
+
+| Rank | Submission | Aggregate | legality | safety | immune_risk | No-fabrication |
+|---|---|---|---|---|---|---|
+| — | **pen-stack-reference** (anchor) | 1.00 | 1.00 | 1.00 | 1.00 | ✓ |
+
+*The reference (PEN-STACK itself) is the leaderboard anchor — it answers from the same validated layers the
+held-out labels come from, so it scores 1.00 by construction. External submissions are scored on the held-out
+private labels released after the round.*
+
+**Standing adoption criterion (the honest non-code bottleneck since v3.1):** ≥1 external integration + ≥1
+external submission. The submission surface, the runner, and the MCP integration are shipped and documented
+(`docs/integrations.md`); landing an external participant depends on outreach. This file is updated as external
+submissions arrive.
+
+To submit, see [README.md](README.md) and [`docs/integrations.md`](../../docs/integrations.md).

pen_stack-5.13.0/docs/autonomy.md

@@ -0,0 +1,36 @@
+# Autonomy levels — PEN-STACK stops at Level 3 (v5.12, WS-AUTONOMY)
+
+PEN-STACK's closed loop (v5.12) is a **Level-3** design→build→test→learn engine: closed, but with **humans/lab in
+control at every gate**. This is the program's deliberate, honest ceiling — not Level 5.
+
+## The levels (for self-driving labs)
+
+| Level | Description | PEN-STACK |
+|---|---|---|
+| 0 | Manual — a human does everything | — |
+| 1 | Assisted — tools score/check, human decides each step | v3.3 verifier |
+| 2 | Partial — the system proposes designs + experiments; human runs + decides | v5.8–v5.11 |
+| **3** | **Closed loop with human control at every gate** — the system runs the full cycle, but pauses for human approval at safety, build, and belief-admission, and flags anomalies | **v5.12 (here)** |
+| 4 | Supervised autonomy — runs many cycles with human oversight, escalating exceptions | not claimed |
+| 5 | Full autonomy — no human in the loop | **not claimed; not a goal** |
+
+## What "Level 3" means here (asserted criteria)
+
+1. **Closed loop.** One command runs generate → predict → decide → safety → build → ingest → learn end-to-end
+   (`run_loop`), in silico (sim-lab) or with a real lab at the same interface.
+2. **Human in control at every gate.** The loop pauses for the `approver` at **safety** (export is refused for a
+   flagged design), **build** (protocols are DRAFTs requiring human/lab review), and **belief-admission** (results
+   enter the curated world-model only via the v4.5 gate with explicit approval).
+3. **Anomaly flagging.** Drift between predictions and observations is detected and **widens uncertainty** rather
+   than over-trusting a stale model; continual updates are versioned + reversible.
+4. **No fabrication.** Every number is tool-sourced; no stage invents a value.
+
+## What PEN-STACK deliberately does NOT do
+
+- It does **not** run experiments autonomously (protocols are drafts; a human/lab runs them).
+- It does **not** auto-edit the curated world-model (admission requires human approval).
+- It does **not** retrain the foundation models (continual learning is recalibration).
+- It does **not** claim Level 4 or 5. The human is in the loop by design.
+
+> Level 3 is a trustworthy ceiling: a closed, gated, drift-aware loop that a scientist drives — not an autonomous
+> agent acting in the world. The convergence demonstration is retrospective/simulated, reported with CIs.

pen_stack-5.13.0/docs/challenge.md

@@ -0,0 +1,33 @@
+# The Genome-Writing Challenge (v5.13)
+
+PEN-STACK's accumulated bench tasks (v3.4→v5.12) become an **open, recurring, held-out** benchmark others build
+*to* — the CASP / Virtual-Cell-Challenge model for the *writing* side of genome engineering.
+
+```bash
+python benchmarks/genome_writing_challenge/run.py          # score the PEN-STACK reference on the current round
+```
+
+## How it works (`benchmarks/genome_writing_challenge/harness.py`)
+
+An external agent submits a `Submission(name, predict_fn)`. `evaluate(submission, round_id)` scores
+`predict_fn(public_input) -> answer` on a **held-out** round:
+
+- each `public_input` names its `family` + `task_id` + `design` + `instructions` — **never the label**;
+- the private label is computed by the **validated PEN-STACK layers** (the v3.3 rules, the v5.7 Guardian, the
+  v5.6 immune profile) — so **no task uses a circular label**; a fabricated answer cannot match an invented label
+  and simply scores 0;
+- a **no-fabrication** audit runs on every submission;
+- task families generalise the internal bench: `legality`, `safety`, and an **immune-risk** task grounded in the
+  v5.6 oracles (future rounds add write-types, adversarial, outcome, experiment-design, closed-loop).
+
+The **reference** submission (PEN-STACK itself) anchors the leaderboard at 1.0 by construction.
+
+## Why it is honest
+
+- **Held-out + private labels** released after a round — you cannot reverse-fit.
+- **No circular labels** — labels are mechanistic/verifier facts, not the agent's own claim.
+- **Reproducible** — deterministic scoring, version-pinned.
+- **Immune-risk first-class** — an immune-risk task draws directly on the v5.6 profile.
+
+See [`benchmarks/genome_writing_challenge/README.md`](../benchmarks/genome_writing_challenge/README.md),
+`SUBMISSIONS.md`, and [Integrations](integrations.md).

pen_stack-5.13.0/docs/closed_loop.md

@@ -0,0 +1,54 @@
+# The closed loop (v5.12)
+
+From v5.12, PEN-STACK integrates every prior cycle into one continual **design→build→test→learn** loop — humans/lab
+in control at every gate, no fabrication, drift-aware. One command runs it end-to-end.
+
+```python
+from pen_stack.loop import run_loop
+result = run_loop(goal, cell_state="k562", candidates=pool, rounds=5, approver="human")
+result["autonomy_level"]      # 3
+result["human_in_control"]    # True
+result["history"]             # per-round: n, blocked, best_readout, drift, versioned update
+```
+
+## The DBTL orchestrator (`pen_stack/loop/cycle.py`)
+
+Each round composes the whole stack:
+
+1. **generate** (v5.8) — safe + legal + calibrated + immune-profiled candidates (the verifier-as-discriminator
+   discards hazardous/illegal proposals);
+2. **decide** (v5.10) — `select_batch` picks a diverse, informative batch (EIG + immune-VOI);
+3. **safety + build** (v5.7 + v5.11) — `export_protocol` is **safety-gated** (a flagged design is blocked);
+4. **test** (v5.11) — `run_simulated` (sim-lab) or a real lab at the same interface;
+5. **ingest** (v4.5) — results enter as candidate evidence, admitted only through the gate;
+6. **drift** (v5.12) — predicted vs observed;
+7. **learn** (v5.12) — `continual_update` recalibrates, versioned + reversible.
+
+The loop **pauses for the approver at safety, build, and belief-admission** — it is not autonomous.
+
+## Drift detection (`pen_stack/loop/drift.py`)
+
+`detect_drift(designs, results)` compares the twin's predictions against observed readouts. Growing
+miscalibration → `severity: "high"` → `action: "inflate_intervals"` — widen uncertainty rather than over-trust a
+stale model. Covers calibration/residual shift, not every failure mode.
+
+## Continual learning (`pen_stack/loop/continual.py`)
+
+`continual_update(admitted_results, drift=…, approver=…, prev_version=…)` recalibrates the trust layer, the v5.9
+twin, and the v5.6 immune proxies **on admitted outcomes only**. Every update is **versioned** (evidence digest)
+and **reversible** (`rollback_to` the prior version), attributable to the approver. High drift widens intervals.
+An admitted immune measurement **with a CI** can move a v5.6 axis proxy → outcome-validated. This is
+**recalibration, not foundation-model retraining**.
+
+## Convergence demonstration
+
+`loop_converges_faster_than_random` reports — retrospectively, with a bootstrap CI — whether the loop's active
+Learn stage reaches a target model quality in fewer rounds than random selection (the v5.10 validation). It is
+reported honestly either way; the demonstration is retrospective/simulated.
+
+## Honest scope
+
+The loop is **Level 3** — closed, but with humans/lab in control at every gate, **not autonomous**. It runs in
+silico via the sim-lab (a real lab attaches at the same interface); continual learning recalibrates rather than
+retrains; drift detection covers calibration/residual shift, not all failures; immune-proxy graduation requires an
+admitted measurement with a CI. See [Autonomy levels](autonomy.md).

pen_stack-5.13.0/docs/co_scientist_loop.md

@@ -0,0 +1,31 @@
+# The co-scientist over the loop (v5.13)
+
+The matured co-scientist drives the whole Series-II loop for a working scientist — every output safe, legal,
+calibrated, cited, scope-ledgered, and **immune-profiled**, and never fabricated. **The co-scientist drives and
+presents; the scientist/lab decides.**
+
+```python
+from pen_stack.agent.co_scientist import co_scientist_session
+session = co_scientist_session(goal, cell_state="k562")
+```
+
+Returns, for a documented goal:
+
+| Key | From | What |
+|---|---|---|
+| `strategies` | v5.8 | the Pareto frontier of designs (incl. the grounded immune-risk axis) |
+| `predicted_outcomes` | v5.9 | calibrated outcomes with intervals + scope flags |
+| `immune_profiles` | v5.6 | the per-axis immune-risk vector, **first-class** (`collapsed_score is None`) |
+| `suggested_experiments` | v5.10 | the diverse, informative batch to run next (EIG + immune-VOI) |
+| `protocols_available` | v5.11 | safety-gated protocol export on request (DRAFT) |
+| `citations` | v5.0 | a literature-cited rationale (citations resolve by construction) |
+| `scope_ledger` | v5.0 | what was assessed vs not (the known-unknowns made legible) |
+| `safety` | v5.7 | the per-design safety decision (cleared / flagged) |
+
+Hazardous candidates are discarded by the safety-gated pipeline before they ever appear. No number is
+fabricated; the immune-risk profile is presented **with its known-unknowns**, never as a patient prediction.
+
+## Honest scope
+
+The co-scientist runs the loop and **presents options** — it does not decide. Safety, no-fabrication, calibration,
+and the scope ledger hold throughout. It is the most useful *face* of the substrate, not an autonomous agent.